Download Ants on the Move: Resource Limitation of a Litter

BIOTROPICA 38(3): 419–427 2006
10.1111/j.1744-7429.2006.00153.x
Ants on the Move: Resource Limitation of a Litter-nesting Ant Community in Costa Rica1
Terrence P. McGlynn2
Department of Biology, University of San Diego, 5998 Alcalá Park, San Diego, California 92110, U.S.A.
ABSTRACT
The leaf litter of tropical wet forests is replete with itinerant ant nests. Nest movement may help ants evade the constraints of stress and disturbance and increase access
to resources. I studied how nest relocation and environmental factors may explain the density, size, and growth of leaf litter ant nests. I decoupled the relationships
among litter depth, food abundance, and nest availability in a 4-mo manipulation of food and leaf litter in a community of litter-nesting ants in a lowland wet forest
in Costa Rica. Over 4 mo, 290 1 m2 treatment and control plots were sampled without replacement. Nest densities doubled in response to food supplementation, but
did not decrease in response to litter removal or stress (from litter trampling). The supplementation of food increased the utilization of less favored nesting materials. In
response to food supplementation and litter trampling, arboreal ants established nests in the litter, and growth rates of the most common ants (Pheidole spp.) increased.
Colony growth was independent of colony size and growth rates of the most abundant ants. In general, I conclude that litter-nesting ant density is driven primarily
by food limitation, that nest relocation behavior significantly affects access to resource and the demographic structure of this community, and that nest fission may be
a method to break the growth–reproduction trade-off.
RESUMEN
La hojarasca de bosques húmedos tropicales está repleta de nidos itinerantes de hormiga. El movimiento de nido puede ayudar a las hormigas a evadir las limitaciones
producidas por tensión y perturbación y a la vez aumentar el acceso a recursos. lLas relaciones entre la profundidad de la hojarasca, la abundancia de alimento y la
disponibilidad de nido fueron desconectadas durante 4 meses debido a la manipulación de hojarasca y de alimento en un bosque húmedo bajo en Costa Rica. Durante
4 meses, 290 parcelas de tratamiento y control de 1 m2 fueron muestreadas sin reemplazo. Las densidades de nido doblaron en respuesta al alimento suplementario,
pero no disminuyeron en respuesta a la eliminación de la hojarasca y la tensión (causada al pisotear la hojarasca). El alimento suplementario aumentó la utilización de
materiales menos favorecidos para anidar. Debido al alimento suplementario y la tensión en la hojarasca, las hormigas arbóreas establecieron nidos en la hojarasca y las
tasas de crecimiento de las hormigas más comunes (Pheidole spp.) aumentaron. El crecimiento de la colonia era independiente del tamaño de la colonia, y de las tasas
de crecimiento de las hormigas más abundantes. En terminos generales, se concluye que la densidad de las hormigas en la hojarasca es gobernada primordialmente por
la limitación de alimento, que la conducta de transladar nido afecta significativamente el acceso a recursos y a la estructura demográfica de esta comunidad, y que la
fisión de nido puede ser un método para romper el intercambio entre crecimiento y reproducción.
Key words: competition; Costa Rica; density; leaf litter; nest relocation; Pheidole; resource limitation; Solenopsis; tropical rain forest.
IN TROPICAL RAIN FORESTS, nest relocation is a regular part of the
life history of most ant species, particularly those nesting within
leaf litter (Byrne 1994). When resource quality declines, ants may
move their nests to patches of higher resource abundance or reduced
densities of competitors. Because litter is highly patchy within the
scale of ant home ranges, one might expect that nest relocation may
be an important mediator of competition (McGlynn et al. 2004).
To date, little is known about the resources limiting litter ants,
mainly because they move their nests (but see Kaspari 1996a,b).
Food, nests, or territories are limiting for many species (Carroll &
Janzen 1973, Ryti & Case 1992, Sanders & Gordon 2003), often observed through competitive interactions among species (Savolainen
& Vespalainen 1988, Porter & Savignano 1990, Andersen & Patel
1994, Adams & Tschinkel 1995, Tschinkel & Hess 1999, Sanders
et al. 2003). It is difficult to determine which resources are limiting
(Andersen 1995, Bestelmeyer & Wiens 1996, Morrison 1996) and
hence determine density and demography of ant nests (Fernandez
Escudero & Tinaut 1999, Brown & Gordon 2000, McGlynn et al.
2002).
1 Received
18 February 2005; revision accepted 28 June 2005.
author; e-mail: [email protected]
2 Corresponding
In a community of ants that moves nests frequently, what
regulates density and growth? Within the leaf litter environment,
potential regulators include nest availability, food abundance, food
distribution, and microclimate (Kaspari 1996a, Banschbach et al.
1997, McGlynn & Kirksey 2000, Durou et al. 2001). All of these
variables are in turn mediated by the depth of leaf litter, which
provides nests, retains humidity, and forms the foundation of the
detritial food web (Moore et al. 2004). Manipulative experiments
can decouple the natural correlations among litter abundance, food
abundance, and microclimate (Levings & Windsor 1984). Even
though field researchers cannot alter microclimate and natural food
resources in situ without confounding environmental changes, manipulations of resources and microhabitat nevertheless will provide
new insights into the environmental factors that limit leaf litter
ants.
In this study, I use experimental manipulations to assess how
food and leaf litter affect the density and demography of ants that
move their nests within the leaf litter of a tropical wet forest. I
created four alterations of food or leaf litter at the spatial scale of a
single square meter to be sampled over 4 mo. Because ants in this
community move their nests on the scale of weeks, nest relocation
may be inferred by increases or decreases in nest density.
!
C 2006 The Author(s)
C 2006 by The Association for Tropical Biology and Conservation
Journal compilation !
419
420
McGlynn
METHODS
This study was conducted from January to June 1997 in an old
growth forest at La Selva Biological Station, Sarapiquı́ Canton,
Heredia Province, Costa Rica. La Selva is a lowland wet tropical
forest in the Caribbean Lowlands of Costa Rica, and receives ca
4 m of rain annually, with the majority falling in the wet season
from June to December (McDade & Hartshorn 1994). The 4-ha
site was near the intersection of the Camino Circular Cercano and
Sendero Suroeste on the La Selva trail system. Using the La Selva
GIS, the site was bounded by 1100 and 1300 on the 32◦ axis, and
600 and 800 on the 122◦ axis. I demarcated 290 plots of 1 m2
size using wire flags and plastic flagging. These plot locations were
at least 5 m apart and at least 20 m away from trails. In similar
habitat at the same field station, Kaspari (1996b) found that 1 m2
plots placed 5 m apart from one another were far enough to ensure
independent sampling of leaf litter-nesting ant species. Plots were
randomly assigned in equal proportion among the control and four
treatment categories.
I created four treatments altering food abundance, food distribution, and leaf litter quantity and quality. I refer to these treatments as clumped food, split food, removal, and trampling. Both
litter treatments caused a reduction in the depth of leaf litter, but
trampling caused the destruction of potential nests and litter compaction, and removal caused a drop in prey abundance without
reduction in the number of nests. Each of the food treatments received approximately 5 g (about 20 individuals) of Nasutitermes
corniger termites, applied every 2 d. Termites were collected directly
from arboreal carton nests at La Selva, and then frozen in a −20◦ C
freezer. Termites were lightly coated with vegetable oil to attract a
wide variety of foraging ants in the leaf litter community. In the
clumped food plots, termites were placed in a single pile in an arbitrarily selected location inside a 20-cm circle located at the center of
the plot. In the split food treatments, the same quantity of food was
distributed equally among four quarters of the plot. The leaf litter
removal treatment was designed to increase stress, through removal
of prey biomass and reducing humidity. I applied this treatment
once, at the beginning of the experiment; I removed approximately
three-fourths of the leaf litter from the ground, without removing any ant nests. I only removed leaf matter, without removing
any twigs, sticks, nuts, seeds, or seedpods. Any leaves containing
ant nests were replaced to the plot. The trampling plots received a
weekly treatment, stepping firmly ten times in rubber wading boots,
size 10–11 (US). When a research assistant with size 7 (US) boots
stepped in the plots, she stepped 12 times, to result in an equivalent
surface area subjected to trampling.
The experiment was designed to prevent any disturbance or
stress upon the field colonies. Consequently, plots were sampled
continuously without replacement, with collections distributed over
the course of 4 mo after the start of treatments. Approximately
every 2 d, one plot per treatment was sampled. I collected leaf litter
ant nests from the plots using the “intensive sampling” protocol
described by Bestelmeyer et al. (2000). In the field, all matter resting
upon the ground inside the plot was collected and brought into the
laboratory in plastic bags. Immediately following collection, I sorted
TABLE 1.
Frequency of common nest materials among treatment plots. Nest material rarely used within control plots was used more frequently by ants
in the food supplementation plots (χ 2 = 47.18, df = 28, P = 0.013).
Clumped Split
Nest material
Control
food
food Removal Trampled Total
Diffuse
Leaf
Palm leaf or petiole
3
6
2
10
25
11
11
25
12
8
14
4
8
9
9
40
79
38
Pentaclethra
macroloba Pod
Hard seed
3
13
15
0
6
37
3
2
11
6
4
26
17
61
74
32
36
220
Stick
Twig
Wood
72
134
12
78
172
15
75
175
20
48
114
13
73
142
12
346
737
72
Subtotal: woody
nests
218
265
270
175
227
1155
Total
235
326
344
207
263
1375
Subtotal: nonwoody
nests
through the leaf litter breaking every single twig, stick, nut, piece of
wood, panacle, bark, and seedpod to ensure that every single litterdwelling ant was collected. In nearly all cases, nests remained intact
and all of the contents were collected into a single vial. Throughout
the first month since the treatments started, I collected 80 plots,
equally distributed among five treatment categories. During the
second, third, and fourth months, I collected 85, 70, and 55 plots,
respectively, for a total of 290 plots.
Nest materials were sorted into several categories (Table 1).
“Diffuse” nests were not located in a discrete cavity and may have
been enclosed between two types of materials or exposed to the
surface. Leaf nests were found inside curled up leaves or between
two leaves. Nests on the surface of the common dehisced pods of
Pentaclethra macroloba (Mimosaceae) were relatively exposed compared to other nests. “Hard seeds” represented a variety of plant
taxa and were not identified. Twigs had a diameter of <1 cm, while
sticks were thicker than twigs. Pieces of “wood” were noncylindrical
pieces of decomposing wood that were neither sticks nor twigs. I
excluded 74 nests from the contingency analysis of nest materials.
The excluded nest categories included those that were so decomposed that the materials were not identifiable, and infrequently used
materials, such as infructescences of the palm Welfia regia and the
fruit capsules of Apeiba membranacea (Tiliaceae).
Nests were brought to the laboratory for sorting, identification,
and counting of ants. I counted the number of workers, worker
pupae, dealate queens, winged queens, pupal queens, males, pupal males, and larvae. For the dimorphic genus Pheidole, I distinguished between major and minor workers and pupae; these results
are presented elsewhere (McGlynn & Owen 2002). I identified every ant to species or morphospecies, according to Bolton (1994),
Longino and Hanson (1995), Longino and Cover (pers. comm.
Limitation of Litter-Nesting Ants
TABLE 2.
Functional
group
Nest frequencies per treatment sorted by functional group. Functional
group composition was significantly heterogeneous among control and
treatment plots (χ 2 = 61.89, df = 24, P < 0.0001).
Control
Clumped
food
Split
food
Removal
Trampled
Total
0
40
6
51
5
43
0
37
5
35
16
206
Paratrechina
Pheidole
Solenopsis
2
114
47
16
151
47
20
164
74
6
89
46
8
128
76
52
646
290
Wasmannia
Others
8
28
19
54
20
39
12
36
2
22
61
179
239
344
365
226
276
1450
Arboreal
Dacetine
Total
and http://www.evergreen.edu/ants), and Wilson (2003). I use the
nest as the unit of analysis in this study, because colony boundaries
could not be reliably determined by species-level identification or
by queen number.
Because of high species richness, no species was sufficiently
abundant (despite the collection of 1450 nests overall) to allow for
meaningful species-level comparisons among the treatments, even
though there were 58 plots per treatment. To facilitate comparisons of community members, I grouped species together on the
bases of taxonomy and functional role in the community (Table 2).
Such “functional groups” are parallel to those of Andersen (1995),
Bestelmeyer and Wiens (1996), and McGlynn (1999a), and represent a reasonable approach to evaluating components of a “hyperdiverse” community. Here I list the criteria for designating functional groups. “Arboreal” consists of species that are known mainly
from sampling of canopy ants and have been rarely or never observed within leaf litter, according to sampling of the Arthropods of
La Selva project (www.evergreen.edu/ants). Arboreal ants included
Azteca spp., Crematogaster spp., and Pheidole christopherseni. Members of the genera Pyramica and Strumigenys constitute the “Dacetine” group; they are generally believed to be specialized predators of
arthropod microfauna (Hölldobler & Wilson 1990, Dejean 1991).
The two abundant genera Pheidole and Solenopsis were each designated to their own functional groups; little is known about their
natural history and there was little basis for analyzing these genera by species groups. Two species, sole representatives of their
genus in this study, have been assigned to their own groups: Paratrechina steinheili and Wasmannia auropunctata. The foraging ecology of these two species differs from the other litter-nesting species.
P. steinheili is an opportunist that does not defend food resources
to which it recruits (McGlynn & Kirksey 2000). W. auropunctata
has a diffuse colony structure and is an invasive species in other
parts of the world (LeBreton et al. 2003). Some taxa were relatively
uncommon or not primarily litter-nesting species and were simply
excluded from analyses other than those of total nest density. These
species included: two Camponotus, one Adelomyrmex, one Megalomyrmex, two Gnamptogenys, one Anochetus, two Discothyrea, four
421
Cyphomyrmex, one Cerapachys, and over two dozen Hypoponera nests
that could not be completely collected. A complete list is available
in another publication (McGlynn 1999b).
After counting the contents of each identified nest, I measured
demographic variables for each colony. Colony size is defined by the
number of adult workers. Growth may be defined by the ratio of pupal workers to adult workers (as in Michener 1964, Kaspari 1996b,
McGlynn & Owen 2002, McGlynn et al. 2004). Reproduction is
the sum of pupae and adults of reproductive castes.
Statistical tests were informed by the use of Dytham’s dichotomous key (1999). I tested whether nesting habits changed in response to treatments by comparing the frequency of nest types
among the treatments using a chi-square test. I tested the effects of
treatments upon nest size, growth, and reproduction using analysis of variance (ANOVA). The demographic responses were tested
for time dependent effects using two-way ANOVAs with month of
treatment duration as the second factor, using separate analyses for
each functional group. The Arboreal and Wasmannia groups have
been excluded from several analyses because sample sizes were not
adequate for the appropriate statistical tests.
To rule out the possibility that increases in nest density were
purely the result of nest fission instead of nest relocation, I performed
correlation analyses for each treatment, comparing the number of
nests in each plot with the mean number of workers per nest. If
changes in nest density were only attributable to nest fission, then
nest fission events would simultaneously increase nest density and
decrease nest sizes, resulting in a negative relationship between nest
density and colony size. I tested whether the density of nests in each
plot was independent among functional groups; one would expect
negative associations if one functional group displaced another. Because the distribution of nest frequencies was not normal, I created
used rank correlations. To explain additional variance in the density
of ant nests, I evaluated the effects of leaf litter depth upon nest
density using linear regression. To determine how the treatments
affected nest growth rates, I performed polynomial regression on
the number of worker pupae relative to the number of workers.
RESULTS
NEST MATERIAL.—Over the course of the experiment, 1450 nests
were collected from control and treatment plots. The frequencies
of nest material significantly differed among treatments. Marginal
nest materials that rarely were occupied in the control plots often
contained ant nests in the food supplementations (Table 1).
NEST DENSITY.—Over 4 mo, both food supplementation treatments
caused an increase of the number of nests, compared to the control
plots, while removal and trampled treatments showed no change
relative to the controls (Fig. 1). A two-way ANOVA on nest density
demonstrated significant effects of treatment (F 4280 = 5.31, P <
0.001), month of treatment duration (F 4280 = 3.95, P < 0.01), and
an interaction between month and treatment (F 4280 = 1.80, P <
0.05). To determine whether the increase in density was caused by
increases in particular functional groups, I examined the changes in
422
McGlynn
TABLE 3.
Comparisons of nest density per plot, using Kendall’s rank-order correlation coefficients, T, ∗ P < 0.05 after Bonferroni correction for multiple
comparisons.
Functional
group
Arboreal
Dacetine
Paratrechina
Pheidole
Solenopsis
Wasmannia
Arboreal Dacetine Paratrechina Pheidole Solenopsis Wasmannia
–
−0.002
–
0.006
0.016
0.015
0.021
0.184∗
0.136∗
–
0.046
0.066∗
–
0.171
–
∗
−0.001
−0.017
0.019∗
−0.025
−0.006
–
I found no association between nest size and nest density;
the highest R2 value was 0.026, indicating that the changes in the
number of nests could not have resulted solely from fragmentation.
Functional groups did not apparently displace one another, as nest
density among functional groups was independent of one another;
the greatest tau value was 0.18 (Table 3). Most associations were
weakly positive, which was likely caused by a general increase in
nest density with deeper leaf litter. I found that leaf litter depth did
not affect nest density in the control plots. Leaf litter depth was
positively correlated with nest density in three of the treatments,
but not the clumped food (Fig. 2).
FIGURE 1.
The response of nest density (per m2 ) to treatments applied to
m2
1
plots. Open circles indicate control plots and error bars represent ±1
standard error.
nest density among groups (Table 2). The increase in nest density
occurred in both food treatments among Arboreal, Paratrechina,
Pheidole, and Wasmannia. None of these groups changed in nest
density with trampling. Solenopsis, on the other hand, experienced
increased nest densities in the split food and trampling treatments,
but not in the clumped food treatment.
DEMOGRAPHIC RESPONSES TO TREATMENTS.—I compared demographic responses to the treatments using nest size (number of
workers), growth rate (worker brood/workers), and reproduction
(number of reproductives) as metrics. Table 4 presents the effects
of treatments on the nest demographic parameters in each functional group. The treatments caused significant effects on the size,
growth, or reproduction of Pheidole nests only. There were no effects
of month of treatment duration or interacting effects between the
demographic factors and the treatment duration.
In all of the treatments, among all the four functional groups
considered, the maximum colony size was larger than in the controls. In the large nests of Paratrechina, Pheidole, and Solenopsis, the
shapes of the curves indicated greater growth in the split food treatments compared to the clumped food treatments, often driven by a
few outlying observations (Fig. 3). In many cases, large colonies continued to grow as quickly, if not more quickly, than small colonies.
FIGURE 2. The relationship between the density of litter ant nests and the depth of leaf litter in 1 m2 plots, among control and treatment plots. Linear regression,
∗∗ P < 0.01; ∗∗∗ P < 0.001.
Limitation of Litter-Nesting Ants
TABLE 4.
423
Demographic responses to treatments among functional groups. ANOVA comparisons were performed within each functional group between each treatment and
control; ∗ P < 0.05 and ∗∗ P < 0.01 after Bonferroni correction. Two-way ANOVA results in every functional group indicated no interacting effects of treatment
duration, using four categories designated for the months of treatment duration.
Functional
group
Arboreal
Treatment
Control
Clumped food
Split food
Removal
Trampled
Dacetine
Control
Clumped food
Split food
Removal
Trampled
Paratrechina
Pheidole
Control
Clumped food
Split food
Removal
Trampled
Control
Clumped food
Split food
Removal
Solenopsis
Trampled
Control
Clumped food
Split food
Removal
Wasmannia
Trampled
Control
Clumped food
Split food
Removal
Trampled
Number of adults,
mean ± SE (N)
(0)
57.8 ± 26.1 (6)
103.6 ± 46.0 (5)
(0)
5.8 ± 3.2 (5)
13.7 ± 1.5 (40)
13.5 ± 2.7 (51)
20.1 ± 4.2 (43)
23.35 ± 5.4 (37)
17.0 ± 2.7 (35)
Alates, mean ± SE (N)
Growth, mean ± SE (N)
(0)
0 ± 0 (6)
(0)
0.23 ± 0.15 (6)
1.85 ± 0.68 (40)
1.43 ± 0.66 (51)
0.40 ± 0.06 (37)
0.45 ± 0.11 (48)
0.6 ± 0.4 (5)
(0)
0 ± 0 (5)
1.81 ± 0.56 (43)
1.65 ± 0.46 (37)
0.80 ± 0.28 (35)
0.54 ± 0.22 (5)
(0)
0 ± 0 (3)
0.23 ± 0.05 (41)
0.31 ± 0.08 (32)
0.30 ± 0.05 (33)
13 ± 4.0 (2)
28.7 ± 5.3 (16)
20 ± 3.6 (20)
0 ± 0 (2)
0.19 ± 0.10 (16)
2.10 ± 0.93 (20)
54.7 ± 4.7 (151)
62.9 ± 5.0 (164)∗
41.9 ± 5.6 (89)
2.52 ± 0.62 (151)∗
2.87 ± 0.73 (164)∗∗
1.10 ± 0.39 (89)
0.44 ± 0.06 (151)∗
0.33 ± 0.03 (163)
0.46 ± 0.07 (85)∗
30.2 ± 4.4 (74)
26.9 ± 2.5 (46)
1.10 ± 0.83 (74)
0.44 ± 0.23 (46)
0.37 ± 0.05 (74)
0.55 ± 0.12 (45)
105.6 ± 27.8 (20)
72.6 ± 16.1 (12)
237.0 ± 231.0 (2)
0.25 ± 0.18 (20)
0 ± 0 (12)
0 ± 0 (2)
27.5 ± 8.8 (6)
36.5 ± 12.7 (8)
46.4 ± 4.1 (114)
40.6 ± 4.3 (128)
29.2 ± 5.2 (47)
30.7 ± 5.2 (47)
30.9 ± 3.2 (76)
80.3 ± 55.6 (8)
84.2 ± 18.0 (19)
The majority of nests did not produce reproductive caste members (Table 5). In Pheidole, one of few groups where reproductive
caste members were present, the frequency of the production of
reproductive caste members was significantly greater than in the
controls.
DISCUSSION
NEST RELOCATION.—The increase in nest density of the food supplementation treatments was primarily caused by nest relocation.
We can eliminate nest fission as a major cause of the increase in density because nest size did not decline with increases in nest density.
It is not possible to know whether ants relocated outside the plots
during the treatment period, but it is clear that the net influx of ants
into these plots was greater than the net rate of departure. Nests in
0 ± 0 (6)
1.5 ± 0.93 (8)
0.73 ± 0.32 (114)
1.31 ± 0.46 (128)
0.04 ± 0.3 (47)
0.11 ± 0.09 (47)
0.37 ± 0.17 (76)
0.50 ± 0.50 (8)
0 ± 0 (19)
0.14 ± 0.08 (2)
0.28 ± 0.07 (16)
0.18 ± 0.06 (20)
0.20 ± 0.08 (6)
0.28 ± 0.09 (8)
0.27 ± 0.03 (112)
0.33 ± 0.4 (125)
0.37 ± 0.05 (47)
0.40 ± 0.05 (47)
0.43 ± 0.05 (76)
0.70 ± 0.49 (8)
0.58 ± 0.28 (19)
0.54 ± 0.30 (20)
0.12 ± 0.07 (12)
0 ± 0 (2)
food supplementation plots may choose to move less frequently, or
move short distances to remain within the plot. Alternatively, nests
that relocate may be more likely to select the areas subjected to food
treatment.
LIMITATION OF NEST DENSITY.—It appears that both food and nests
can be limiting. When food is limiting, nest sites are not limiting.
When food is added, nest sites appear to become the limiting factor.
Two lines of evidence support this hypothesis. First, when food was
added, ants shifted to suboptimal nesting habitats. Second, leaf litter
depth and nest density positively correlated in the supplemental
food treatments but not in the control.
Because nest availability and food availability are often coupled
in nature (even if loosely), it would have been difficult to evaluate
these relationships without supplemental food. Even though relocation brings more ants into areas with supplemental food, there is
424
McGlynn
FIGURE 3.
< 0.001.
Worker–worker pupae curves measure investment into growth as nest sizes increase. Second order polynomial regression, ∗ P < 0.05; ∗∗ P < 0.01; ∗∗∗ P
no evidence that ants emigrate away from bad conditions or poor
food quality. There was no decline in the nest density in the litter
(and hence food) removal treatment, or in the trampling treatment
which resulted in the destruction of many potential nests. Considering that the depth of leaf litter is closely associated with the
number of available nests, the release of food constraint may then
reveal nest constraint. In the trampling treatment, resulting in the
destruction of some potential nests, there was no decline in nest
density, although litter depth was correlated with nest density.
If nests are limiting, then the apparent abundance of potential and unoccupied nests begs explanation. Many potentially high
quality nests are unoccupied at any given moment. While picking
through leaf litter, ants often are found nesting in the most paltry
locations while hollow sticks and twigs remain unoccupied. In another study, when supplemental nests were added to the litter, nest
densities were found to increase even though few of the supplemental nests were occupied (Kaspari 1996a). It appears as if the existence
of empty nests may have value, even if they are unoccupied.
I suggest that the value of unoccupied nests is a function of
the frequent relocation of litter-nesting species. Although nest relocations are frequent in the litter, no selective force driving a regular
pattern of nest relocation has yet to be determined in any species
(McGlynn et al. 2004). Nest relocation could occur to improve
access to resources or avoid competition, because relocations often
place nests in more favorable locations (Smallwood 1982). However, at least one ant species at La Selva, the common ground-nesting
Aphaenogaster araneoides, apparently relocates its nests about once
per week independent of resource availability and environmental
conditions (McGlynn et al. 2004). Aphaenogaster araenoides may
move in response to army ants (McGlynn et al. 2004). Under the
army ant predation hypothesis, ant nests accumulate odors, which
attract army ants. It is assumed that army ants use short-distance
olfactory cues to locate prey, though no such experiment has been
conducted (Gotwald 1995). Even though there is no direct evidence
for the army ant predation hypothesis yet, other hypotheses are not
supported and the circumstantial evidence is intriguing.
GROWTH AND REPRODUCTION.—Social insects are typically thought
to resolve the trade-off between growth and reproduction by initially
investing in growth when nests are small, then shifting to reproductive castes when more workers are in the nest (Oster & Wilson
1978). In most social insect nests, which grow and reproduce in
modular units, this switch in energetic allocation is easily evaluated
by counting the number of members of each caste (Bourke & Franks
1995). Surprisingly, in the present study, there is no evidence that
ants defer reproduction until reaching a threshold nest size.
Kaspari (1996b) proposed that the independence of nest size
and relative allocation toward sterile and reproductive castes is
caused by a constant rate of mortality. According to this argument,
because death might arrive at any minute, selection does not favor
a delay in the production of reproductive caste members. I offer an
alternative explanation for the fixed allocation to reproductive and
sterile castes: namely, that frequent colony fission breaks the tradeoff between growth and reproduction. At least some and perhaps
Limitation of Litter-Nesting Ants
TABLE 5.
Percentage of nests containing reproductive caste members. Chi-square values were calculated on frequency values used to determine significant differences among
treatments within functional groups. The observed frequency values in each category are in parentheses. df for all tests is 12.
Functional
Percent
reproductive
Percent
Percent
Percent
castes absent
only males
only females
both sexes
N
χ2
P
10.47
0.57
12.01
0.45
12.13
0.049
7.90
0.79
group
Treatment
Dacetine
Control
Clumped food
Split food
70.0 (28)
76.5 (39)
53.5 (23)
5.0 (2)
2.0 (1)
9.3 (4)
12.5 (5)
13.7 (7)
18.6 (8)
12.5 (5)
7.8 (4)
18.6 (8)
40
51
43
Removal
Trampled
64.9 (24)
74.3 (26)
2.7 (1)
2.7 (2)
13.5 (5)
14.3 (5)
18.9 (7)
2.7 (2)
37
35
Paratrechina
Pheidole
Solenopsis
425
100 (2)
0
0
0
2
Clumped food
Split food
Removal
Control
81.3 (13)
70.0 (14)
100 (6)
12.5 (2)
10.0 (2)
0
6.3 (1)
15.0 (3)
0
0
10.0 (2)
0
16
20
6
Trampled
62.5 (5)
0
37.5 (3)
0
Control
82.5 (94)
7.0 (8)
7.9 (9)
2.6 (3)
114
Clumped food
Split food
Removal
69.5 (105)
68.3 (112)
84.3 (75)
13.2 (20)
14.0 (23)
5.6 (5)
14.6 (22)
11.6 (19)
7.9 (7)
2.6 (4)
6.1 (10)
2.2 (2)
151
164
89
Trampled
78.9 (101)
12.5 (16)
7.0 (9)
1.6 (2)
128
Control
Clumped food
95.7 (45)
95.7 (45)
2.1 (1)
0
2.1 (1)
2.1 (1)
0
2.1 (1)
47
47
Split food
Removal
89.2 (66)
89.1 (41)
2.7 (2)
0
4.1 (3)
8.7 (4)
4.1 (3)
2.2 (1)
74
46
Trampled
82.1 (70)
1.3 (1)
3.9 (3)
2.6 (2)
76
most litter ant species can reproduce by fission. Colonies of animals
that reproduce via fission have broken the growth–reproduction
trade-off, by allocating growth units toward reproduction at the
time of fission. It is hard to estimate rates of fission in the field,
because of the difficulty differentiating fission and relocation events
(Peeters & Ito 2001). Sterile workers are a prerequisite for fission,
and the number of successful fission events is tied to the number of
sterile workers produced within a nest. As a result of translating nest
growth into reproductive potential, fission-reproducing ants greatly
restrict the range of dispersal from the natal nest.
Changes in the distribution of colony size with food supplementation offer indirect evidence as to how fission rate and relocation relate. In plots where food was supplemented, the maximum
colony size consistently increased. While maximum colony size may
be a function of the size at which colonies die, it is more likely to
be a function of the size at which colonies split. Thus, the upper
limit on observed nest sizes may be determined by the behavioral
decision to split one nest into two. This decision to split a nest is
probably based on foraging efficiency (e.g., Garb et al. 2000). At
the size at which foraging efficiency declines with increasing colony
size, colonies can be expected to split (they do not just slow their
growth since we know that growth rate is independent of colony
size). The greater nest sizes in the food supplementation treatments
may thus demonstrate that colonies have decided to split at a larger
threshold size due to a higher rate of payoff for larger colonies when
resources are plentiful.
8
Whereas work at other sites makes clear predictions as to how
allocation to reproduction varies with colony size, predictions for
how growth rate (worker/pupae curves) scales with colony size come
largely from other work in the tropics. Two aspects of colony growth
rate can be compared, the mean growth rate (mean worker/pupae
ratio) and the relationship between growth rate and colony size.
Our results apparently contradict “Michener’s paradox”: that per
capita worker production declines with the number of adult workers
in the nest (Michener 1964). Kaspari (1996a) found that colony
growth rates are generally independent of colony size. In the present
study, within the stress and disturbance treatments, colony growth
rates were greater or equal to those in the controls. In the food
supplementation plots, colony growth rates increased as colonies
grew in the split food plots, but nests in the clumped food plots
showed slower growth in large colonies. Large nests in the clumped
food plots slow down their growth. The opposite pattern occurred
for the food supplementation for the dacetines, but this makes
sense as the dacetines are probably specialized predators who were
competing for nest space with the other functional groups, but not
for the supplemental food.
NESTING HABITS OF THE LITTER ANT COMMUNITY.—Changes in
the abundance of food in the litter created several changes in nesting behavior. In both food treatments, ants were found within nest
materials that were rarely occupied in the control plots. Moreover,
nests of arboreal species increased in density. Whereas arboreal ants
426
McGlynn
are occasionally found at traps and baits on the ground (McGlynn
& Kirksey 2000, T. McGlynn, pers. obs.), these ants were thought
to access the litter while foraging from aboveground nests. Because
arboreal ants are probably nitrogen-limited (Davidson et al. 2003),
then a predictable source of nitrogen-rich food apparently is adequate to entice some ants to nest within the litter.
ACKNOWLEDGMENTS
Mark Springer volunteered 5 mo of fieldwork. Several undergraduates assisted with the field experiment and colony processing, including Jeb Owen, Fielding Arnold, Bobby Bascomb, Erica Brady, Jess
Burlason, Rob Ickes, Candice Kelley, Eben Kirksey, Sarah Lamond,
and Lloyd Rowland. Mike Breed, Jeb Owen, Nate Sanders, Bonnie
Ripley, and two anonymous reviewers helped with the interpretation of results and editing of the manuscript, and Rob Dunn was
instrumental in its completion. Emily Barker translated the abstract
into Spanish. This project was funded by a University of San Diego
Faculty Research Grant, an Organization for Tropical Studies Dissertation Grant, the Andrew Mellon Foundation, and Amelia.
LITERATURE CITED
ADAMS, E. S., AND W. R. TSCHINKEL. 1995. Spatial dynamics of colony interactions in young populations of the fire ant Solenopsis invicta. Oecologia
102: 156–163.
ANDERSEN, A. N. 1995. A classification of Australian ant communities, based
on functional groups which parallel plant life-forms in relation to stress
and disturbance. J. Biogeogr. 22: 15–29.
———, AND A. D. PATEL. 1994. Meat ants as dominant members of Australian
ant communities: An experimental test of their influence on the foraging success and forager abundance of other species. Oecologia 98: 15–
24.
BANSCHBACH, V. S., N. LEVIT, AND J. M. HERBERS. 1997. Nest temperatures
and thermal preferences of a forest ant species: Is seasonal polydomy a
thermoregulatory mechanism? Insectes Soc. 44: 109–122.
BESTELMEYER, B. T., D. AGOSTI, L. ALSONSO, C. R. BRANDÃO, W. L. BROWN,
JR., J. H. C. DELABIE, AND R. SILVESTRE. 2000. Field techniques for the
study of ground-dwelling ants: An overview, description and evaluation.
In D. Agosti, L. Alonso, J. Majer, and T. Schultz (Eds.). Ants: Standard methods for measuring and monitoring biodiversity, pp. 122–144.
Smithsonian Institution Press, Washington, District of Columbia.
———, AND J. A. WIENS. 1996. The effects of land use on the structure of
ground-foraging ant communities in the Argentine chaco. Ecol. Appl.
6: 1225–1240.
BOLTON, B. 1994. Identification guide to ant genera of the world. Harvard
University Press, Cambridge, Massachusetts.
BOURKE, A. F. G., AND N. R. FRANKS. 1995. Social evolution in ants, p. 529.
Princeton University Press, Princeton, New Jersey.
BROWN, M. J. F., AND D. M. GORDON. 2000. How resources and encounters
affect the distribution of foraging activity in a seed-harvesting ant. Behav.
Ecol. Sociobiol. 47: 195–203.
BYRNE, M. M. 1994. Ecology of twig-dwelling ants in a wet lowland tropical
forest. Biotropica 26: 61–72.
CARROLL, C. R., AND D. H. JANZEN. 1973. Ecology of foraging by ants. Annu.
Rev. Ecol. Syst. 4: 231–257.
DAVIDSON, D. W., S. C. COOK, R. R. SNELLING, AND T. H. CHUA. 2003.
Explaining the abundance of ants in lowland tropical rainforest canopies.
Science 300: 969–972.
DEJEAN, A. 1991. Gathering of nectar and exploitation of Aphididae by
Smithistruma emarginata (Formicidae: Myrmicinae). Biotropica 23:
207–208.
DUROU, S., J. LAUGA, AND A. DEJEAN. 2001. Intensive food searching in humid
patches: Adaptation of a myrmicinae ant to environmental constraints.
Behaviour 138: 251–259.
DYTHAM, C. 1999. Choosing and using statistics: A biologist’s guide. Blackwell
Science, Oxford.
FERNÁNDEZ ESCUDERO, I., AND A. TINAUT. 1999. Factors determining nest distribution in the high-mountain ant Proformica longiseta (Hymenoptera:
Formicidae). Ethol. Ecol. Evol. 11: 325–338.
GARB, J., B. P. KOTLER, AND J. S. BROWN. 2000. Foraging and community
consequences of seed size for coexisting Negev Desert granivores. Oikos
88: 291–300.
GOTWALD, W. H. 1995. Army ants: The biology of social predation. Comstock
Publishing Associates, Ithaca, New York.
HÖLLDOBLER, B., AND E. O. WILSON. 1990. The ants. Belknap Press, Cambridge, Massachusetts.
KASPARI, M. 1996a. Testing resource-based models of patchiness in four
Neotropical litter ant assemblages. Oikos 76: 443–454.
———. 1996b. Litter ant patchiness at the 1-m2 scale: Disturbance dynamics
in three Neotropical forests. Oecologia 107: 265–273.
LEBRETON, J., J. CHAZEAU, AND H. JOURDAN. 2003. Immediate impacts of
invasion by Wasmannia auropunctata (Hymenoptera: Formicidae) on
native litter ant fauna in a New Caledonian rainforest. Aust. Ecol. 28:
204–209.
LEVINGS, S. C., AND D. M. WINDSOR. 1984. Litter moisture content as a
determinant of litter arthropod distribution and abundance during the
dry season on Barro Colorado Island, Panama. Biotropica 16: 125–
131.
LONGINO, J. T., AND P. E. HANSON. 1995. The ants (Formicidae). In P. E.
Hanson and I. D. Gauld (Eds.). The Hymenoptera of Costa Rica, pp.
588–620. Oxford University Press, London.
MCDADE, L. A., AND G. S. HARTSHORN. 1994. La Selva biological station. In
L. A. McDade, K. S. BAWA, H. A. Hespenheide, and G. S. Hartshorn
(Eds.). La Selva: Ecology and natural history of a Neotropical rain forest,
pp. 6–14. University of Chicago Press, Chicago, Illinois.
MCGLYNN, T. P. 1999a. The worldwide transport of ants: Geographic distribution and ecological invasions. J. Biogeogr. 26: 535–548.
———. 1999b. The biogeography, behavior, and ecology of exotic ants. Ph.D.
dissertation, University of Colorado, Boulder, Colorado.
———, R. CARR, J. CARSON, AND J. BUMA. 2004. Frequent nest relocation in
the ant Aphaenogaster araneoides: Resources, competition, and natural
enemies. Oikos 106: 611–621.
———, J. R. HOOVER, G. S. JASPER, M. S. KELLY, A. M. POLIS, C. M. SPANGLER,
AND B. J. WATSON. 2002. Environmental heterogeneity affects demography of the Central American ant Aphaenogaster araneoides. J. Trop. Ecol.
18: 231–244.
———, AND S. E. KIRKSEY. 2000. The effect of food presentation and microhabitat upon resource monopoly in a ground-foraging ant (Hymenoptera:
Formicidae) community. Rev. Biol. Trop. 48: 629–642.
———, AND J. P. OWEN. 2002. Food supplementation alters caste allocation in
a natural population of Pheidole flavens, a dimorphic leaf-litter dwelling
ant. Insectes Soc. 49: 8–14.
MICHENER, C. D. 1964. Reproductive efficiency in relation to colony size in
hymenopterous societies. Insectes Soc. 11: 317–341.
MOORE, J. C., E. L. BERLOW, D. C. COLEMAN, P. C. DE RUITER, Q. DONG,
A. HASTINGS, N. COLLINS JOHNSON, K. S. MCCANN, K. MELVILLE, P. J.
MORIN, K. NADELHOFFER, A. D. ROSEMOND, D. M. POST, J. L. SABO,
K. M. SCOW, M. J. VANNI, AND D. H. WALL. 2004. Detritus, trophic
dynamics and biodiversity. Ecol. Lett. 7: 584–600.
MORRISON, L. W. 1996. Community organization in a recently assembled fauna:
The case of Polynesian ants. Oecologia 107: 243–256.
OSTER, G. F., AND E. O. WILSON. 1978. Caste and ecology in the social insects.
Princeton University Press, Princeton, New Jersey.
Limitation of Litter-Nesting Ants
PEETERS, C., AND F. ITO. 2001. Colony dispersal and the evolution of queen
morphology in social Hymenoptera. Annu. Rev. Entomol. 46: 601–
630.
PORTER, S. D., AND D. A. SAVIGNANO. 1990. Invasion of polygyne fire ants
decimates native ants and disrupts arthropod community. Ecology 71:
2095–2106.
RYTI, R. T., AND T. J. CASE. 1992. The role of neighborhood competition
in the spacing and diversity of ant communities. Am. Nat. 139: 355–
374.
SANDERS, N. J., AND D. M. GORDON. 2003. Resource-dependent interactions
and the organization of desert ant communities. Ecology 84: 1024–1031.
427
———, N. J. GOTELLI, N. E. HELLER, AND D. M. GORDON. 2003. Community
disassembly by an invasive ant species. Proc. Natl. Acad. Sci. USA 100:
2474–2477.
SAVOLAINEN, R., AND K. VESPALAINEN. 1988. A competition hierarchy among
boreal ants: Impact on resource partitioning and community structure.
Oikos 51: 135–155.
SMALLWOOD, J. 1982. Nest relocations in ants. Insectes Soc. 29: 138–147.
TSCHINKEL, W. R., AND C. A. HESS. 1999. Arboreal ant community of a pine
forest in northern Florida. Ann. Entomol. Soc. Am. 92: 63–70.
WILSON, E. O. 2003. Pheidole in the new world: A dominant, hyperdiverse ant
genus. Harvard University Press, Cambridge, Massachusetts.